# Forum Feedback **URL:** https://forum.langchain.com/c/forum-feedback/2.md [Latest](https://forum.langchain.com/latest.md) · [Categories](https://forum.langchain.com/categories.md) · [Tags](https://forum.langchain.com/tags.md) --- ## [About the Forum Feedback category](https://forum.langchain.com/t/about-the-forum-feedback-category/1)
**Author:** [@system](https://forum.langchain.com/u/system)\ **Replies:** 0\ **Last updated:** [April 22, 2025, 6:59pm UTC](https://forum.langchain.com/t/about-the-forum-feedback-category/1 "2025-04-22T18:59:15Z")
Discussion about our Help Forums, its organization, how it works, and how we can improve it. --- ## [Does langchain provides classes for multimodal output models or for simplicity models with image only output like GPT-Image-2? Is the ChatOpenAI() going to work for GPT-Image-2 model?](https://forum.langchain.com/t/does-langchain-provides-classes-for-multimodal-output-models-or-for-simplicity-models-with-image-only-output-like-gpt-image-2-is-the-chatopenai-going-to-work-for-gpt-image-2-model/4698)
**Author:** [@shubhamkala06](https://forum.langchain.com/u/shubhamkala06)\ **Replies:** 4\ **Last updated:** [October 7, 2026, 6:35am UTC](https://forum.langchain.com/t/does-langchain-provides-classes-for-multimodal-output-models-or-for-simplicity-models-with-image-only-output-like-gpt-image-2-is-the-chatopenai-going-to-work-for-gpt-image-2-model/4698 "2026-10-07T06:35:35Z")
I have been trying to figure out that is it possible that a model response can be a image block or any other block other than the text block. So can a model response give an image block like this- { "type": "image… --- ## [Add serverless-workflow library: compile CNCF Serverless Workflow DSL into LangGraph StateGraph](https://forum.langchain.com/t/add-serverless-workflow-library-compile-cncf-serverless-workflow-dsl-into-langgraph-stategraph/4267)
**Author:** [@hambuger](https://forum.langchain.com/u/hambuger)\ **Replies:** 4\ **Last updated:** [July 31, 2026, 3:24am UTC](https://forum.langchain.com/t/add-serverless-workflow-library-compile-cncf-serverless-workflow-dsl-into-langgraph-stategraph/4267 "2026-07-31T03:24:33Z")
Proposal Add a new serverless-workflow library under libs/ that compiles a CNCF Serverless Workflow ( GitHub - open-workflow-specification/specification: Contains the official specification for the Open Workflow Specific… --- ## [How are you enforcing hard spending limits on agents in production?](https://forum.langchain.com/t/how-are-you-enforcing-hard-spending-limits-on-agents-in-production/3894)
**Author:** [@DeboJolaosho](https://forum.langchain.com/u/DeboJolaosho)\ **Replies:** 2\ **Last updated:** [July 29, 2026, 2:36pm UTC](https://forum.langchain.com/t/how-are-you-enforcing-hard-spending-limits-on-agents-in-production/3894 "2026-07-29T14:36:36Z")
How are you all handling cost enforcement on agents — not just monitoring but a hard gate before tool calls fire? Working on something for this and curious what patterns people are using. --- ## [Should agent memory/tools account for source reliability, not just relevance?](https://forum.langchain.com/t/should-agent-memory-tools-account-for-source-reliability-not-just-relevance/4204)
**Author:** [@falinapterus](https://forum.langchain.com/u/falinapterus)\ **Replies:** 4\ **Last updated:** [July 28, 2026, 2:38pm UTC](https://forum.langchain.com/t/should-agent-memory-tools-account-for-source-reliability-not-just-relevance/4204 "2026-07-28T14:38:48Z")
Hi all — I come from a database reliability background (Postgres/MySQL, currently at TigerData/Timescale), and I’ve been building TekmerDB — an open-source (Apache 2.0) memory/storage layer for AI agents that does someth… --- ## [Support to initialize StateGraph with instance of the class](https://forum.langchain.com/t/support-to-initialize-stategraph-with-instance-of-the-class/4226)
**Author:** [@justOleh](https://forum.langchain.com/u/justOleh)\ **Replies:** 4\ **Last updated:** [July 23, 2026, 9:40am UTC](https://forum.langchain.com/t/support-to-initialize-stategraph-with-instance-of-the-class/4226 "2026-07-23T09:40:34Z")
Hi there! I’m taking course - AI Agents in LangGraph - DeepLearning.AI where Harrison Chase explains how to build AI agent using langgraph. In his implementation he is doing the following: class Agent: def \_\_init\_\_(se… --- ## [Statefold – Event-sourced state, replay and time travel for LangGraph agents](https://forum.langchain.com/t/statefold-event-sourced-state-replay-and-time-travel-for-langgraph-agents/4187)
**Author:** [@Joshuaakaspace](https://forum.langchain.com/u/Joshuaakaspace)\ **Replies:** 0\ **Last updated:** [July 19, 2026, 12:01pm UTC](https://forum.langchain.com/t/statefold-event-sourced-state-replay-and-time-travel-for-langgraph-agents/4187 "2026-07-19T12:01:29Z")
Hi everyone, I’m sharing Statefold, an open-source, framework-agnostic state and observability platform for AI agents. GitHub: GitHub - ioteverythin/statefold: Framework-agnostic, event-sourced state platform for AI ag… --- ## [Should AI agents expose their reasoning process before taking action?](https://forum.langchain.com/t/should-ai-agents-expose-their-reasoning-process-before-taking-action/4131)
**Author:** [@Saiganesh](https://forum.langchain.com/u/Saiganesh)\ **Replies:** 2\ **Last updated:** [July 10, 2026, 7:45pm UTC](https://forum.langchain.com/t/should-ai-agents-expose-their-reasoning-process-before-taking-action/4131 "2026-07-10T19:45:18Z")
Humans build trust by explaining why they made a decision. What if AI agents briefly summarized their plan before executing? For example: “I found three scheduling conflicts. I’ll resolve the first automatically, ask … --- ## [Proposal: composable ToolNode middleware (tool-call dedup + wrap\_tool\_call chaining)](https://forum.langchain.com/t/proposal-composable-toolnode-middleware-tool-call-dedup-wrap-tool-call-chaining/4098)
**Author:** [@zamaniali1995](https://forum.langchain.com/u/zamaniali1995)\ **Replies:** 2\ **Last updated:** [July 7, 2026, 8:27pm UTC](https://forum.langchain.com/t/proposal-composable-toolnode-middleware-tool-call-dedup-wrap-tool-call-chaining/4098 "2026-07-07T20:27:10Z")
I opened PR #8291 to add a few small, dependency-free utilities to langgraph.prebuilt for two patterns that teams frequently re-implement when working with ToolNode: build\_tool\_call\_key, deduplicate\_tool\_calls, and de… --- ## [Loop prevention in Deep Agents with repeated tool calls](https://forum.langchain.com/t/loop-prevention-in-deep-agents-with-repeated-tool-calls/4034)
**Author:** [@Jewells](https://forum.langchain.com/u/Jewells)\ **Replies:** 1\ **Last updated:** [June 26, 2026, 10:27am UTC](https://forum.langchain.com/t/loop-prevention-in-deep-agents-with-repeated-tool-calls/4034 "2026-06-26T10:27:38Z")
Loop Prevention Middleware in Deep Agents I’m using a custom loop prevention middleware that tracks the tool parameters and responses. If the same tool is called multiple times with the same parameters and produces the s… --- ## [LangGraph orchestration Problem](https://forum.langchain.com/t/langgraph-orchestration-problem/4017)
**Author:** [@akin](https://forum.langchain.com/u/akin)\ **Replies:** 5\ **Last updated:** [June 24, 2026, 9:00am UTC](https://forum.langchain.com/t/langgraph-orchestration-problem/4017 "2026-06-24T09:00:30Z")
Guys, I have a problem. I’m building an orchestration layer, and the agent—for example—asks the user for a keyword, extracts the keyword, but it also calls the Excel tool and writes the data to Excel. The user didn’t ask… --- ## [\[Architecture Feedback\] Multi-agent orchestration system for solo TPE — LangGraph + Nextcloud](https://forum.langchain.com/t/architecture-feedback-multi-agent-orchestration-system-for-solo-tpe-langgraph-nextcloud/3930)
**Author:** [@Olivier](https://forum.langchain.com/u/Olivier)\ **Replies:** 0\ **Last updated:** [June 15, 2026, 1:27pm UTC](https://forum.langchain.com/t/architecture-feedback-multi-agent-orchestration-system-for-solo-tpe-langgraph-nextcloud/3930 "2026-06-15T13:27:59Z")
\*\*Description\*\* I’m a solo entrepreneur running a renovation company (Brittany, France). I’m designing a multi-agent orchestration system to automate my business processes across 6 areas (admin/finance, commercial, tech… --- ## [Response Format Groq Model Pydantic](https://forum.langchain.com/t/response-format-groq-model-pydantic/3843)
**Author:** [@Galangzz](https://forum.langchain.com/u/Galangzz)\ **Replies:** 1\ **Last updated:** [June 2, 2026, 12:30pm UTC](https://forum.langchain.com/t/response-format-groq-model-pydantic/3843 "2026-06-02T12:30:30Z")
i try to build some ai agent with groq model \`meta-llama/llama-4-scout-17b-16e-instruct\`. i want to structured the output with pydantic. But, it cant. ❯ python main.py Enter your query: analisis saham BBCA dong Tracebac… --- ## [Best Stack for Building AI Applications](https://forum.langchain.com/t/best-stack-for-building-ai-applications/3690)
**Author:** [@aartijangid](https://forum.langchain.com/u/aartijangid)\ **Replies:** 0\ **Last updated:** [May 8, 2026, 1:13pm UTC](https://forum.langchain.com/t/best-stack-for-building-ai-applications/3690 "2026-05-08T13:13:55Z")
What is the best modern stack for building scalable AI applications in 2026? Discuss: LLMs (GPT, Claude, Gemini, Llama) Frameworks (LangChain, LangGraph, LlamaIndex) RAG pipelines & vector databases AI agents and wo… --- ## [agentrial: statistical testing for LangGraph agents (open source)](https://forum.langchain.com/t/agentrial-statistical-testing-for-langgraph-agents-open-source/2940)
**Author:** [@alepot55](https://forum.langchain.com/u/alepot55)\ **Replies:** 1\ **Last updated:** [February 12, 2026, 10:08pm UTC](https://forum.langchain.com/t/agentrial-statistical-testing-for-langgraph-agents-open-source/2940 "2026-02-12T22:08:04Z")
Hey everyone, I built a CLI tool for testing LangGraph agents with statistical rigor. Sharing here because I think it might be useful for others dealing with the “my agent works sometimes” problem. \*\*The problem I was … --- ## [OpenAI API client isnide creat\_agent insted of ChatOpenAI](https://forum.langchain.com/t/openai-api-client-isnide-creat-agent-insted-of-chatopenai/2891)
**Author:** [@Petar](https://forum.langchain.com/u/Petar)\ **Replies:** 1\ **Last updated:** [February 5, 2026, 4:23pm UTC](https://forum.langchain.com/t/openai-api-client-isnide-creat-agent-insted-of-chatopenai/2891 "2026-02-05T16:23:51Z")
Is it possible to use the OpenAI API directly without using the LangChain wrapper for create\_agent in LangGraph, and also trace this OpenAI client in LangSmith traces (usage, metadata, etc.)? Insted of using this: retu… --- ## [Issues with the Usage of Thinking Patterns](https://forum.langchain.com/t/issues-with-the-usage-of-thinking-patterns/2621)
**Author:** [@ryh114514](https://forum.langchain.com/u/ryh114514)\ **Replies:** 0\ **Last updated:** [December 30, 2025, 6:50am UTC](https://forum.langchain.com/t/issues-with-the-usage-of-thinking-patterns/2621 "2025-12-30T06:50:44Z")
When using the invoke method of chatdeepseek, I always remove the reason field, which prevents the return of tool calls. The transmitted information is as follows:\[{‘role’: ‘user’, ‘content’: ‘小米16怎么样’}, {‘role’: ‘assist… --- ## [What happened to github /discussions/?](https://forum.langchain.com/t/what-happened-to-github-discussions/2310)
**Author:** [@gabrii](https://forum.langchain.com/u/gabrii)\ **Replies:** 2\ **Last updated:** [November 23, 2025, 6:12pm UTC](https://forum.langchain.com/t/what-happened-to-github-discussions/2310 "2025-11-23T18:12:39Z")
I understand the change to this forum, nothing wrong with that. This topic is about the disappearance of the github discussions on the langchain repository. As someone new to langchain, I’m googling multiple issues and … --- ## [When contributing, where in Langchain Forum do I ask for assignment?](https://forum.langchain.com/t/when-contributing-where-in-langchain-forum-do-i-ask-for-assignment/1861)
**Author:** [@seho0808](https://forum.langchain.com/u/seho0808)\ **Replies:** 0\ **Last updated:** [October 17, 2025, 3:00pm UTC](https://forum.langchain.com/t/when-contributing-where-in-langchain-forum-do-i-ask-for-assignment/1861 "2025-10-17T15:00:07Z")
I am thinking of implementing some retreivers, documen loaders, tools, etc that exist in python but not in JS version. (for python i found all of the libraries are in the langchain community repo, but for JS it was in li… --- ## [Cloud Architecture Diagram Generator](https://forum.langchain.com/t/cloud-architecture-diagram-generator/1393)
**Author:** [@oscarismael47](https://forum.langchain.com/u/oscarismael47)\ **Replies:** 0\ **Last updated:** [August 28, 2025, 6:30pm UTC](https://forum.langchain.com/t/cloud-architecture-diagram-generator/1393 "2025-08-28T18:30:57Z")
:rocket: Excited to share my latest personal project: Diagram Generator! :framed\_picture: I built a Streamlit-based web app that lets you generate and visualize architecture diagrams (like AWS) just by describing them i… --- ## [Introduction😞 way to complicated for an uninitiated](https://forum.langchain.com/t/introduction-way-to-complicated-for-an-uninitiated/1194)
**Author:** [@Bitrios](https://forum.langchain.com/u/Bitrios)\ **Replies:** 0\ **Last updated:** [August 16, 2025, 3:23pm UTC](https://forum.langchain.com/t/introduction-way-to-complicated-for-an-uninitiated/1194 "2025-08-16T15:23:56Z")
Just signed up and I’ve so far read through the getting setup section and am completely stumped. It’s not to say that I can’t eventually complete the setup but undoubtedly it’s the work of days compared to what an intro… --- ## [No "Unread" tab at the top?](https://forum.langchain.com/t/no-unread-tab-at-the-top/210)
**Author:** [@DavidSchargel](https://forum.langchain.com/u/DavidSchargel)\ **Replies:** 4\ **Last updated:** [July 12, 2025, 3:19pm UTC](https://forum.langchain.com/t/no-unread-tab-at-the-top/210 "2025-07-12T15:19:58Z")
On other Discourse forums, I can dismiss a whole bunch of messages by clicking on the “Unread” tab and then the “Dismiss” button. I don’t see it on this forum, however. --- ## [Dark mode please!](https://forum.langchain.com/t/dark-mode-please/108)
**Author:** [@mdrxy](https://forum.langchain.com/u/mdrxy)\ **Replies:** 1\ **Last updated:** [July 4, 2025, 9:19pm UTC](https://forum.langchain.com/t/dark-mode-please/108 "2025-07-04T21:19:23Z")
Our eyes would appreciate it :full\_moon\_face: :new\_moon\_face: --- ## [THANK YOU for this forum!](https://forum.langchain.com/t/thank-you-for-this-forum/80)
**Author:** [@TonyG](https://forum.langchain.com/u/TonyG)\ **Replies:** 1\ **Last updated:** [July 1, 2025, 10:49pm UTC](https://forum.langchain.com/t/thank-you-for-this-forum/80 "2025-07-01T22:49:32Z")
I believe this forum will be really good for the community. Please: Advertise it. Direct people here. Build the community. Watch https://meta.discourse.org/ for platform changes. Monitor for technical and social/commun… --- ## [Increase the number of links allowed?](https://forum.langchain.com/t/increase-the-number-of-links-allowed/168)
**Author:** [@edward](https://forum.langchain.com/u/edward)\ **Replies:** 1\ **Last updated:** [July 1, 2025, 10:46pm UTC](https://forum.langchain.com/t/increase-the-number-of-links-allowed/168 "2025-07-01T22:46:40Z")
Can you please increase the number of links allowed from 2- even 3 would make a difference. 2 links is so limiting for providing reference to resources and other questions in the forum.